English

TuneComp: Joint Fine-tuning and Compression for Large Foundation Models

Machine Learning 2025-05-29 v1 Artificial Intelligence

Abstract

To reduce model size during post-training, compression methods, including knowledge distillation, low-rank approximation, and pruning, are often applied after fine-tuning the model. However, sequential fine-tuning and compression sacrifices performance, while creating a larger than necessary model as an intermediate step. In this work, we aim to reduce this gap, by directly constructing a smaller model while guided by the downstream task. We propose to jointly fine-tune and compress the model by gradually distilling it to a pruned low-rank structure. Experiments demonstrate that joint fine-tuning and compression significantly outperforms other sequential compression methods.

Keywords

Cite

@article{arxiv.2505.21835,
  title  = {TuneComp: Joint Fine-tuning and Compression for Large Foundation Models},
  author = {Xiangyu Chen and Jing Liu and Ye Wang and Matthew Brand and Pu and Wang and Toshiaki Koike-Akino},
  journal= {arXiv preprint arXiv:2505.21835},
  year   = {2025}
}

Comments

Preliminary Work

R2 v1 2026-07-01T02:44:52.288Z